docker-compose-2.yml

version: "3.8"

name: ile

networks:
  vnet:
    driver: bridge
    ipam:
      config:
        - subnet: 192.168.130.0/24

volumes:
  redis_data:
    driver: local
    driver_opts:
      o: bind
      type: none
      device: ${PWD}/data_redis

services:
  redis:
    build:
      context: ${PWD}
      no_cache: false
      pull: true
      # https://github.com/docker-library/redis/blob/master/7.0/Dockerfile
      # https://redis.io/docs/management/config-file/
      dockerfile_inline: |
        FROM redis:7.0.12
        RUN mkdir -p /usr/local/etc/redis
        COPY <<EOF /usr/local/etc/redis/redis.conf
        save 3600 1 300 100 60 10000
        appendonly yes
        requirepass ${ILE_REDIS_PASSWORD}
        dir /data
        EOF
        RUN set -eux \
          && groupadd -g "${ILE_NONROOT_GID}" nonroot \
          && useradd -u "${ILE_NONROOT_UID}" -g "${ILE_NONROOT_GID}" nonroot
        RUN chown -R nonroot:nonroot /data
        USER nonroot
        CMD [ "redis-server", "/usr/local/etc/redis/redis.conf" ]
    image: ile-redis:0.0.1
    user: nonroot:nonroot
    restart: always
    hostname: redis
    networks:
      vnet:
        ipv4_address: 192.168.130.20
    volumes:
      - redis_data:/data
    logging:
      options:
        max-size: 10m
        max-file: 1
    healthcheck:
      # https://questdb.io/doc-s/operations/health-monitoring/
      test: timeout -s 15 5 redis-cli -h localhost -p 6379 -a ${ILE_REDIS_PASSWORD} ping | grep -qaFi PONG || exit 1
      interval: 30s
      timeout: 10s
      start_period: 20s
      retries: 3
    deploy:
      restart_policy:
        condition: any
        delay: 5s
      resources:
        limits:
          cpus: '0.4'
        reservations:
          cpus: '0.3'

  tcp-cache:
    build:
      context: ${PWD}/ile_tcp_cache
      args:
        ILE_NONROOT_UID: ${ILE_NONROOT_UID}
        ILE_NONROOT_GID: ${ILE_NONROOT_GID}
      no_cache: false
      pull: true
    image: ile-tcp-cache:0.0.1
    user: nonroot:nonroot
    restart: always
    networks:
      vnet:
        ipv4_address: 192.168.130.21
    extra_hosts:
      - questdb:192.168.130.10
    ports:
      - "9152:9152"
    environment:
      - ITC_MY_TCP_BIND_HOST=0.0.0.0
      - ITC_MY_TCP_BIND_PORT=9152
      - ITC_TARGET_TCP_HOST=${ITC_TARGET_TCP_HOST}
      - ITC_TARGET_TCP_PORT=9009
      - ITC_REDIS_HOST=redis
      - ITC_REDIS_PORT=6379
      - ITC_REDIS_PASSWORD=${ILE_REDIS_PASSWORD}
    logging:
      options:
        max-size: 10m
        max-file: 1
    depends_on:
      redis:
        condition: service_started
        restart: true
    deploy:
      restart_policy:
        condition: any
        delay: 5s
      resources:
        limits:
          cpus: '0.4'
        reservations:
          cpus: '0.3'

  # https://hub.docker.com/_/telegraf
  # https://www.influxdata.com/blog/docker-run-telegraf-as-non-root/
  telegraf2:
    build:
      context: ${PWD}
      no_cache: false
      pull: true
      # https://github.com/influxdata/influxdata-docker/blob/master/telegraf/1.27/Dockerfile
      dockerfile_inline: |
        FROM telegraf:1.27.2
        RUN set -eux \
          && groupadd -g "${ILE_NONROOT_GID}" nonroot \
          && useradd -u "${ILE_NONROOT_UID}" -g "${ILE_NONROOT_GID}" nonroot
        USER nonroot:999
        CMD telegraf --config /etc/telegraf/telegraf.conf --config-directory /etc/telegraf/telegraf.d
    image: ile-telegraf2:0.0.1
    user: nonroot:999
    restart: always
    network_mode: host
    extra_hosts:
      - ile-tcp-cache:192.168.130.21
    environment:
      - HOST_ETC=/hostfs/etc
      - HOST_PROC=/hostfs/proc
      - HOST_SYS=/hostfs/sys
      - HOST_VAR=/hostfs/var
      - HOST_RUN=/hostfs/run
      - HOST_MOUNT_PREFIX=/hostfs
      - LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libnvidia-ml.so
      - ILT_SOCKET_WRITER_ADDRESS=tcp://ile-tcp-cache:9152
      - ILT_HEALTH_SERVICE_ADDRESS=http://127.0.0.1:8653
    volumes:
      - /:/hostfs:ro
      - /var/run/docker.sock:/var/run/docker.sock:ro
      - ${PWD}/ile_telegraf/telegraf.conf:/etc/telegraf/telegraf.conf:ro
      - ${PWD}/ile_telegraf/telegraf-nvidia.conf:/etc/telegraf/telegraf.d/telegraf-nvidia.conf:ro
      - /usr/bin/nvidia-smi:/usr/bin/nvidia-smi:ro
      - /usr/lib/x86_64-linux-gnu/libnvidia-ml.so:/usr/lib/x86_64-linux-gnu/libnvidia-ml.so:ro
    devices:
      - /dev/nvidiactl:/dev/nvidiactl
      - /dev/nvidia0:/dev/nvidia0
    security_opt:
      # https://github.com/influxdata/telegraf/issues/6574
      - apparmor:unconfined
    logging:
      options:
        max-size: 10m
        max-file: 1
    healthcheck:
      # https://github.com/influxdata/telegraf/blob/release-1.27/plugins/outputs/health/README.md
      test: timeout -s 15 5 curl -f http://127.0.0.1:8653 || exit 1
      interval: 30s
      timeout: 10s
      start_period: 20s
      retries: 3
    depends_on:
      tcp-cache:
        condition: service_started
        restart: true
    deploy:
      restart_policy:
        condition: any
        delay: 5s
      resources:
        limits:
          cpus: '0.4'
        reservations:
          cpus: '0.3'